Import Packages¶
In [1]:
import matplotlib.pyplot as plt
import matplotlib.image as mpimg
import numpy as np
import cv2
import glob
import random
import time
from skimage.feature import hog
from sklearn.preprocessing import StandardScaler
from sklearn.svm import LinearSVC
from sklearn.model_selection import train_test_split
%matplotlib inline
In [2]:
def convert_color(image, color_space = 'RGB'):
if color_space != 'RGB':
if color_space == 'HSV':
converted_image = cv2.cvtColor(image, cv2.COLOR_RGB2HSV)
elif color_space == 'LUV':
converted_image = cv2.cvtColor(image, cv2.COLOR_RGB2LUV)
elif color_space == 'HLS':
converted_image = cv2.cvtColor(image, cv2.COLOR_RGB2HLS)
elif color_space == 'YUV':
converted_image = cv2.cvtColor(image, cv2.COLOR_RGB2YUV)
elif color_space == 'YCrCb':
converted_image = cv2.cvtColor(image, cv2.COLOR_RGB2YCrCb)
else: converted_image = np.copy(image)
return converted_image
def get_hog_features(img, orient, pix_per_cell, cell_per_block,
vis=False, feature_vec=True):
# Call with two outputs if vis==True
if vis == True:
features, hog_image = hog(img, orientations=orient,
pixels_per_cell=(pix_per_cell, pix_per_cell),
cells_per_block=(cell_per_block, cell_per_block),
block_norm= 'L2-Hys',
transform_sqrt=False,
visualise=vis, feature_vector=feature_vec)
return features, hog_image
# Otherwise call with one output
else:
features = hog(img, orientations=orient,
pixels_per_cell=(pix_per_cell, pix_per_cell),
cells_per_block=(cell_per_block, cell_per_block),
block_norm= 'L2-Hys',
transform_sqrt=False,
visualise=vis, feature_vector=feature_vec)
return features
def bin_spatial(img, size=(32, 32)):
color1 = cv2.resize(img[:,:,0], size).ravel()
color2 = cv2.resize(img[:,:,1], size).ravel()
color3 = cv2.resize(img[:,:,2], size).ravel()
return np.hstack((color1, color2, color3))
def color_hist(img, nbins=32): #bins_range=(0, 256)
# Compute the histogram of the color channels separately
channel1_hist = np.histogram(img[:,:,0], bins=nbins)
channel2_hist = np.histogram(img[:,:,1], bins=nbins)
channel3_hist = np.histogram(img[:,:,2], bins=nbins)
# Concatenate the histograms into a single feature vector
hist_features = np.concatenate((channel1_hist[0], channel2_hist[0], channel3_hist[0]))
# Return the individual histograms, bin_centers and feature vector
return hist_features
Load Training Data¶
In [3]:
car_images = sorted(glob.glob('training_images/vehicles/**/*.png'))
noncar_images = sorted(glob.glob('training_images/non-vehicles/**/*.png'))
print(str(len(car_images)) + ' car training data.')
print(str(len(noncar_images)) + ' non-car training data.')
In [4]:
sample_size = 3
sample_car_images = random.sample(car_images, sample_size)
sample_noncar_images = random.sample(noncar_images, sample_size)
fig = plt.figure(figsize = (20, 15))
for i in range(len(sample_car_images)):
car_image = sample_car_images[i]
noncar_image = sample_noncar_images[i]
car_img = mpimg.imread(car_image)
noncar_img = mpimg.imread(noncar_image)
ax1 = fig.add_subplot(3,2,(i*2)+1)
ax1.set_title('Car: ' + car_image)
ax1.imshow(car_img)
ax2 = fig.add_subplot(3,2,(i*2)+2)
ax2.set_title('Non-Car: ' + noncar_image)
ax2.imshow(noncar_img)
HOG¶
Car Samples:¶
In [5]:
color_space = 'YUV' # Can be RGB, HSV, LUV, HLS, YUV, YCrCb
orient = 15 # HOG orientations
pix_per_cell = 8 # HOG pixels per cell
cell_per_block = 2 # HOG cells per block
fig = plt.figure(figsize = (20, 15))
for i in range(len(sample_car_images)):
car_image = sample_car_images[i]
image = mpimg.imread(car_image)
ax1 = fig.add_subplot(3,4,(i*4)+1)
ax1.set_title(car_image + ': Original')
ax1.imshow(image)
converted = convert_color(image, color_space=color_space)
for j in range(3):
features, hog_image = get_hog_features(converted[:,:,j], orient=orient,
pix_per_cell=pix_per_cell, cell_per_block=cell_per_block,
vis=True, feature_vec=False)
ax2 = fig.add_subplot(3,4,(i*4)+j+2)
ax2.set_title('Channel ' + str(j))
ax2.imshow(hog_image, cmap='gray')
Non-Car Samples:¶
In [6]:
color_space = 'YUV' # Can be RGB, HSV, LUV, HLS, YUV, YCrCb
orient = 15 # HOG orientations
pix_per_cell = 8 # HOG pixels per cell
cell_per_block = 2 # HOG cells per block
fig = plt.figure(figsize = (20, 15))
for i in range(len(sample_noncar_images)):
noncar_image = sample_noncar_images[i]
image = mpimg.imread(noncar_image)
ax1 = fig.add_subplot(3,4,(i*4)+1)
ax1.set_title(noncar_image + ': Original')
ax1.imshow(image)
converted = convert_color(image, color_space=color_space)
for j in range(3):
features, hog_image = get_hog_features(converted[:,:,j], orient=orient,
pix_per_cell=pix_per_cell, cell_per_block=cell_per_block,
vis=True, feature_vec=False)
ax2 = fig.add_subplot(3,4,(i*4)+j+2)
ax2.set_title('Channel ' + str(j))
ax2.imshow(hog_image, cmap='gray')
Training Classifier¶
In [7]:
# Define a function to extract features from a list of images
def img_features(feature_image, spatial_feat, hist_feat, hog_feat, hist_bins, orient,
pix_per_cell, cell_per_block, hog_channel):
file_features = []
if spatial_feat == True:
spatial_features = bin_spatial(feature_image, size=spatial_size)
#print 'spat', spatial_features.shape
file_features.append(spatial_features)
if hist_feat == True:
# Apply color_hist()
hist_features = color_hist(feature_image, nbins=hist_bins)
#print 'hist', hist_features.shape
file_features.append(hist_features)
if hog_feat == True:
# Call get_hog_features() with vis=False, feature_vec=True
if hog_channel == 'ALL':
hog_features = []
for channel in range(feature_image.shape[2]):
hog_features.extend(get_hog_features(feature_image[:,:,channel],
orient, pix_per_cell, cell_per_block,
vis=False, feature_vec=True))
hog_features = np.ravel(hog_features)
else:
hog_features = get_hog_features(feature_image[:,:,hog_channel], orient,
pix_per_cell, cell_per_block, vis=False, feature_vec=True)
# Append the new feature vector to the features list
file_features.append(hog_features)
return file_features
def extract_features(imgs, color_space='RGB', spatial_size=(32, 32),
hist_bins=32, orient=9,
pix_per_cell=8, cell_per_block=2, hog_channel=0,
spatial_feat=True, hist_feat=True, hog_feat=True):
# Create a list to append feature vectors to
features = []
# Iterate through the list of images
for file_p in imgs:
file_features = []
image = mpimg.imread(file_p) # Read in each imageone by one
# apply color conversion if other than 'RGB'
if color_space != 'RGB':
if color_space == 'HSV':
feature_image = cv2.cvtColor(image, cv2.COLOR_RGB2HSV)
elif color_space == 'LUV':
feature_image = cv2.cvtColor(image, cv2.COLOR_RGB2LUV)
elif color_space == 'HLS':
feature_image = cv2.cvtColor(image, cv2.COLOR_RGB2HLS)
elif color_space == 'YUV':
feature_image = cv2.cvtColor(image, cv2.COLOR_RGB2YUV)
elif color_space == 'YCrCb':
feature_image = cv2.cvtColor(image, cv2.COLOR_RGB2YCrCb)
else: feature_image = np.copy(image)
file_features = img_features(feature_image, spatial_feat, hist_feat, hog_feat, hist_bins, orient,
pix_per_cell, cell_per_block, hog_channel)
features.append(np.concatenate(file_features))
return features # Return list of feature vectors
In [8]:
# Define parameters for feature extraction
color_space = 'YUV' # Can be RGB, HSV, LUV, HLS, YUV, YCrCb
orient = 15 # HOG orientations
pix_per_cell = 8 # HOG pixels per cell
cell_per_block = 2 # HOG cells per block
hog_channel = 'ALL' # Can be 0, 1, 2, or "ALL"
spatial_size = (32, 32) # Spatial binning dimensions
hist_bins = 32 # Number of histogram bins
spatial_feat = True # Spatial features on or off
hist_feat = True # Histogram features on or off
hog_feat = True # HOG features on or off
car_features = extract_features(car_images, color_space=color_space,
spatial_size=spatial_size, hist_bins=hist_bins,
orient=orient, pix_per_cell=pix_per_cell,
cell_per_block=cell_per_block,
hog_channel=hog_channel, spatial_feat=spatial_feat,
hist_feat=hist_feat, hog_feat=hog_feat)
print('Car samples: ', len(car_features))
noncar_features = extract_features(noncar_images, color_space=color_space,
spatial_size=spatial_size, hist_bins=hist_bins,
orient=orient, pix_per_cell=pix_per_cell,
cell_per_block=cell_per_block,
hog_channel=hog_channel, spatial_feat=spatial_feat,
hist_feat=hist_feat, hog_feat=hog_feat)
print('Non-Car samples: ', len(noncar_features))
X = np.vstack((car_features, noncar_features)).astype(np.float64)
X_scaler = StandardScaler().fit(X) # Fit a per-column scaler
scaled_X = X_scaler.transform(X) # Apply the scaler to X
y = np.hstack((np.ones(len(car_features)), np.zeros(len(noncar_features)))) # Define the labels vector
# Split up data into randomized training and test sets
X_train, X_test, y_train, y_test = train_test_split(scaled_X, y, test_size=0.2, random_state=22)
svc = LinearSVC() # Use a linear SVC
svc.fit(X_train, y_train) # Train the classifier
print('Test Accuracy of SVC = ', round(svc.score(X_test, y_test), 4)) # Check the score of the SVC
Test Images¶
In [9]:
test_images = sorted(glob.glob('test_images/*.jpg'))
print(str(len(test_images)) + ' test images.')
fig = plt.figure(figsize = (20, 10))
for i in range(len(test_images)):
test_image = test_images[i]
image = mpimg.imread(test_image)
ax = fig.add_subplot(2,3,(i)+1)
ax.set_title(test_image)
ax.imshow(image)
In [10]:
# Define a function to draw bounding boxes
def draw_boxes(img, bboxes, color=(0, 0, 255), thick=6):
# Make a copy of the image
imcopy = np.copy(img)
# Iterate through the bounding boxes
for bbox in bboxes:
# Draw a rectangle given bbox coordinates
cv2.rectangle(imcopy, bbox[0], bbox[1], color, thick)
# Return the image copy with boxes drawn
return imcopy
In [11]:
# Define a function that takes an image,
# start and stop positions in both x and y,
# window size (x and y dimensions),
# and overlap fraction (for both x and y)
def slide_window(img, x_start_stop=[None, None], y_start_stop=[None, None],
xy_window=(64, 64), xy_overlap=(0.5, 0.5)):
# If x and/or y start/stop positions not defined, set to image size
if x_start_stop[0] == None:
x_start_stop[0] = 0
if x_start_stop[1] == None:
x_start_stop[1] = img.shape[1]
if y_start_stop[0] == None:
y_start_stop[0] = 0
if y_start_stop[1] == None:
y_start_stop[1] = img.shape[0]
# Compute the span of the region to be searched
xspan = x_start_stop[1] - x_start_stop[0]
yspan = y_start_stop[1] - y_start_stop[0]
# Compute the number of pixels per step in x/y
nx_pix_per_step = np.int(xy_window[0]*(1 - xy_overlap[0]))
ny_pix_per_step = np.int(xy_window[1]*(1 - xy_overlap[1]))
# Compute the number of windows in x/y
nx_buffer = np.int(xy_window[0]*(xy_overlap[0]))
ny_buffer = np.int(xy_window[1]*(xy_overlap[1]))
nx_windows = np.int((xspan-nx_buffer)/nx_pix_per_step)
ny_windows = np.int((yspan-ny_buffer)/ny_pix_per_step)
# Initialize a list to append window positions to
window_list = []
# Loop through finding x and y window positions
# Note: you could vectorize this step, but in practice
# you'll be considering windows one by one with your
# classifier, so looping makes sense
for ys in range(ny_windows):
for xs in range(nx_windows):
# Calculate window position
startx = xs*nx_pix_per_step + x_start_stop[0]
endx = startx + xy_window[0]
starty = ys*ny_pix_per_step + y_start_stop[0]
endy = starty + xy_window[1]
# Append window position to list
window_list.append(((startx, starty), (endx, endy)))
# Return the list of windows
return window_list
In [12]:
# Define a function to extract features from a single image window
# This function is very similar to extract_features()
# just for a single image rather than list of images
def single_img_features(img, color_space='RGB', spatial_size=(32, 32),
hist_bins=32, orient=9,
pix_per_cell=8, cell_per_block=2, hog_channel=0,
spatial_feat=True, hist_feat=True, hog_feat=True):
#1) Define an empty list to receive features
img_features = []
#2) Apply color conversion if other than 'RGB'
if color_space != 'RGB':
if color_space == 'HSV':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2HSV)
elif color_space == 'LUV':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2LUV)
elif color_space == 'HLS':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2HLS)
elif color_space == 'YUV':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2YUV)
elif color_space == 'YCrCb':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2YCrCb)
else: feature_image = np.copy(img)
#3) Compute spatial features if flag is set
if spatial_feat == True:
spatial_features = bin_spatial(feature_image, size=spatial_size)
#4) Append features to list
img_features.append(spatial_features)
#5) Compute histogram features if flag is set
if hist_feat == True:
hist_features = color_hist(feature_image, nbins=hist_bins)
#6) Append features to list
img_features.append(hist_features)
#7) Compute HOG features if flag is set
if hog_feat == True:
if hog_channel == 'ALL':
hog_features = []
for channel in range(feature_image.shape[2]):
hog_features.extend(get_hog_features(feature_image[:,:,channel],
orient, pix_per_cell, cell_per_block,
vis=False, feature_vec=True))
else:
hog_features = get_hog_features(feature_image[:,:,hog_channel], orient,
pix_per_cell, cell_per_block, vis=False, feature_vec=True)
#8) Append features to list
img_features.append(hog_features)
#9) Return concatenated array of features
return np.concatenate(img_features)
# Define a function you will pass an image
# and the list of windows to be searched (output of slide_windows())
def search_windows(img, windows, clf, scaler, color_space='RGB',
spatial_size=(32, 32), hist_bins=32,
hist_range=(0, 256), orient=9,
pix_per_cell=8, cell_per_block=2,
hog_channel=0, spatial_feat=True,
hist_feat=True, hog_feat=True):
#1) Create an empty list to receive positive detection windows
on_windows = []
#2) Iterate over all windows in the list
for window in windows:
#3) Extract the test window from original image
test_img = cv2.resize(img[window[0][1]:window[1][1], window[0][0]:window[1][0]], (64, 64))
#4) Extract features for that window using single_img_features()
features = single_img_features(test_img, color_space=color_space,
spatial_size=spatial_size, hist_bins=hist_bins,
orient=orient, pix_per_cell=pix_per_cell,
cell_per_block=cell_per_block,
hog_channel=hog_channel, spatial_feat=spatial_feat,
hist_feat=hist_feat, hog_feat=hog_feat)
#5) Scale extracted features to be fed to classifier
test_features = scaler.transform(np.array(features).reshape(1, -1))
#6) Predict using your classifier
prediction = clf.predict(test_features)
#7) If positive (prediction == 1) then save the window
if prediction == 1:
on_windows.append(window)
#8) Return windows for positive detections
return on_windows
In [13]:
fig = plt.figure(figsize = (20, 40))
base_size = 64
xy_overlap = (0.75, 0.75)
windows_list = []
windows_list.append([400, 500, 1.0])
windows_list.append([400, 500, 1.2])
windows_list.append([410, 500, 1.4])
windows_list.append([420, 556, 1.6])
windows_list.append([430, 556, 1.8])
windows_list.append([430, 558, 2.0])
windows_list.append([400, 556, 2.2])
windows_list.append([464, 656, 3.0])
windows = []
for i in range(len(windows_list)):
y_start_stop = (windows_list[i][0], windows_list[i][1])
size = int(round(base_size * windows_list[i][2]))
windows += slide_window(image, x_start_stop=[None, None], y_start_stop=y_start_stop,
xy_window=(size,size), xy_overlap=xy_overlap)
for i in range(len(windows_list)):
y_start_stop = (windows_list[i][0], windows_list[i][1])
size = int(round(base_size * windows_list[i][2]))
temp_windows = slide_window(image, x_start_stop=[None, None], y_start_stop=y_start_stop,
xy_window=(size,size), xy_overlap=xy_overlap)
test_image = test_images[0]
image = mpimg.imread(test_image)
draw_image = np.copy(image)
draw_image = draw_boxes(draw_image, temp_windows, color=(0, 0, 255), thick=6)
ax1 = fig.add_subplot(8,2,(i*2)+1)
ax1.set_title(test_image + ': Original')
ax1.imshow(image)
ax2 = fig.add_subplot(8,2,(i*2)+2)
ax2.set_title('Searching Windows - y_start_stop: ' + str(y_start_stop) + ', size: ' + str(size))
ax2.imshow(draw_image)
In [14]:
def find_cars(image):
hot_windows = (search_windows(image, windows, svc, X_scaler, color_space=color_space,
spatial_size=spatial_size, hist_bins=hist_bins,
orient=orient, pix_per_cell=pix_per_cell,
cell_per_block=cell_per_block,
hog_channel=hog_channel, spatial_feat=spatial_feat,
hist_feat=hist_feat, hog_feat=hog_feat))
return hot_windows
In [15]:
fig = plt.figure(figsize = (20, 30))
for i in range(len(test_images)):
test_image = test_images[i]
image = mpimg.imread(test_image)
draw_image = np.copy(image)
# Uncomment the following line if you extracted training
# data from .png images (scaled 0 to 1 by mpimg) and the
# image you are searching is a .jpg (scaled 0 to 255)
image = image.astype(np.float32)/255
hot_windows = find_cars(image)
draw_image = draw_boxes(draw_image, hot_windows, color=(0, 255, 0), thick=6)
ax1 = fig.add_subplot(6,2,(i*2)+1)
ax1.set_title(test_image + ': Original')
ax1.imshow(image)
ax2 = fig.add_subplot(6,2,(i*2)+2)
ax2.set_title('Detected')
ax2.imshow(draw_image)
In [16]:
from scipy.ndimage.measurements import label
def add_heat(heatmap, bbox_list):
# Iterate through list of bboxes
for box in bbox_list:
# Add += 1 for all pixels inside each bbox
# Assuming each "box" takes the form ((x1, y1), (x2, y2))
heatmap[box[0][1]:box[1][1], box[0][0]:box[1][0]] += 1
# Return updated heatmap
return heatmap# Iterate through list of bboxes
def apply_threshold(heatmap, threshold):
# Zero out pixels below the threshold
heatmap[heatmap <= threshold] = 0
# Return thresholded map
return heatmap
def draw_heatmap(img, hot_windows, threshold=1):
heat = np.zeros_like(img[:,:,0]).astype(np.float)
# Add heat to each box in box list
heat = add_heat(heat,hot_windows)
# Apply threshold to help remove false positives
heat = apply_threshold(heat,threshold)
# Visualize the heatmap when displaying
heatmap = np.clip(heat, 0, 255)
return heatmap
def draw_labeled_bboxes(img, labels):
# Iterate through all detected cars
for car_number in range(1, labels[1]+1):
# Find pixels with each car_number label value
nonzero = (labels[0] == car_number).nonzero()
# Identify x and y values of those pixels
nonzeroy = np.array(nonzero[0])
nonzerox = np.array(nonzero[1])
# Define a bounding box based on min/max x and y
bbox = ((np.min(nonzerox), np.min(nonzeroy)), (np.max(nonzerox), np.max(nonzeroy)))
# Draw the box on the image
cv2.rectangle(img, bbox[0], bbox[1], (0,255,0), 6)
# Return the image
return img
def draw_labels(image, heatmap):
# Find final boxes from heatmap using label function
labels = label(heatmap)
draw_img = draw_labeled_bboxes(np.copy(image), labels)
return draw_img
In [17]:
threshold=10
fig = plt.figure(figsize = (20, 30))
for i in range(len(test_images)):
test_image = test_images[i]
image = mpimg.imread(test_image)
draw_image = np.copy(image)
# Uncomment the following line if you extracted training
# data from .png images (scaled 0 to 1 by mpimg) and the
# image you are searching is a .jpg (scaled 0 to 255)
image = image.astype(np.float32)/255
hot_windows = find_cars(image)
heatmap = draw_heatmap(image, hot_windows, threshold=threshold)
labels = draw_labels(draw_image, heatmap)
ax1 = fig.add_subplot(6,3,(i*3)+1)
ax1.set_title(test_image + ': Original')
ax1.imshow(image)
ax2 = fig.add_subplot(6,3,(i*3)+2)
ax2.set_title('Heatmap')
ax2.imshow(heatmap, cmap='hot')
ax3 = fig.add_subplot(6,3,(i*3)+3)
ax3.set_title('Final Outcome')
ax3.imshow(labels)
Video¶
Import Packages¶
In [18]:
from moviepy.editor import VideoFileClip
from IPython.display import HTML
In [19]:
def detect_cars(image):
draw_image = np.copy(image)
# Uncomment the following line if you extracted training
# data from .png images (scaled 0 to 1 by mpimg) and the
# image you are searching is a .jpg (scaled 0 to 255)
image = image.astype(np.float32)/255
hot_windows = find_cars(image)
heatmap = draw_heatmap(image, hot_windows, threshold=threshold)
labels = draw_labels(draw_image, heatmap)
return labels
Test Video¶
In [20]:
threshold=10
test_video_output = 'test_video_output.mp4'
test_video_input = VideoFileClip('test_video.mp4')
processed_test_video = test_video_input.fl_image(detect_cars)
%time processed_test_video.write_videofile(test_video_output, audio=False)
In [21]:
HTML("""
<video width="960" height="540" controls>
<source src="{0}">
</video>
""".format(test_video_output))
Out[21]:
Project Video¶
In [22]:
threshold=10
video_output = 'project_video_out.mp4'
video_input = VideoFileClip('project_video.mp4')
processed_video = video_input.fl_image(detect_cars)
%time processed_video.write_videofile(video_output, audio=False)
In [23]:
HTML("""
<video width="960" height="540" controls>
<source src="{0}">
</video>
""".format(video_output))
Out[23]: